Typically there are multiple routes that a driver may take to reach a destination. Current systems exist that assist a driver in selecting a route to travel to reach a destination. These systems may select, for example, the route with the shortest travel time, that maximizes use of freeways, that minimizes freeways, or that avoids toll roads. One major factor often taken into account by these systems in selecting a route is real-time traffic information, which is information on the current traffic conditions. Typically a driver will want to avoid congested roads.
Real-time traffic information is usually only available for certain roads in urban areas where traffic sensors are placed. For those roads without sensors (e.g., roads in rural areas), the systems assume that the roads are not congested and that vehicles are traveling freely. However, such an assumption may be incorrect and a system may select for the driver to travel on a route that is heavily congested due to an accident, weather conditions or commuter traffic.
Thus, there is a need for a way to be able to predict the traffic conditions of routes for which real-time traffic information is not available.